SAVING PROSPECT LAKE USING PVC

GEOMEMBRANES
By: Stanford W. Slifer,1 Greg J. Monley,2 Kyle Hier,3 and Timothy D. Stark4
Submitted for Review and Possible Publication in the Geotechnical Fabrics Report
September 30, 2005

Prospect Lake has been a landmark in Colorado Springs Memorial Park for more than
100 years and has been the site of many outdoor activities such as swimming, water
skiing, boating, and fishing. However, the severe and prolonged drought in the Colorado
area began to make the estimated annual water loss due to seepage (excluding
evaporation) of about 215 acre-feet/year unacceptable. The 53 acre lake has no natural
recharge and is filled with relatively expensive, treated water. After several years of
drought, the Colorado Springs Department of Parks and Reclamation (CSDPR) realized
that the large seepage loss was politically and economically unsustainable but the public
still wanted to utilize the lake.

Liner System Design

Subsurface conditions encountered within the lake consist of very soft, high plasticity
clay with organics (see Photograph 1) overlying interbedded sands and clays, with
shallow ground water and flowing sands prevalent in the deepest part of the lake. The
lake is also rich in fish hooks, glass and other debris from 100 years of recreational
activity. Initial alternatives for lining the lake included a compacted clay liner (CCL) or
soil admixtures to reduce the permeability of the lakebed soils. The lakebed clay, fish
hooks included, was not suitable borrow for a CCL so importation of clay and
stabilization of the very soft subgrade to allow at least 95% Standard Proctor based
relative compaction (ASTM D 698) of the 2-foot thick CCL would be required. A
similar level of compaction would also be needed for a 2-foot thick layer of soil admixed
lakebed soil, also leading to an unsuitable solution. Difficulties in admixing with the
highly variable lakebed soil also were likely, particularly with the soft, highly plastic
clays prevalent in the deepest part of the lake where providing a low permeability liner is
most critical. Consequently, Kumar and Associates (K+A) recommended that CSDPR
line the entire lake with a geomembrane.

A number of different types of geomembrane were considered but K+A

recommended a 30 mil PVC geomembrane because of cost, ease of installation, long-
term durability, extremely low water vapor transmission, flexibility which results in

1
Vice President, Watersaver Company, Inc., P.O. Box 16465, Denver, CO 80216-0465, 800.525.2424
xt.2403, e-mail: stan.slifer@watersaver.com
2
Senior Project Engineer, Kumar and Associates, 2390 S. Lipan St., Denver, CO 80223, 303-742-9700,
email: gmonley@kumarusa.com
3
President, Scientific Site Construction, Inc., Denver, CO 80223, 303-296-3663, e-mail:
sscinc2@qwest.com,
4
Professor of Civil and Environmental Engineering, University of Illinois, 205 N. Mathews Ave., Urbana,
IL 61801, 217-333-7394, e-mail: tstark@uiuc.edu ,

1
significant puncture resistance, and impressive elongation which provides a high
tolerance to differential settlement. The puncture resistance is important because great
quantities of glass and fish hooks were interspersed throughout the lake bed soils. The
differential settlement capability is important because of the many areas of soft clay or
lake bed sediment, which could continue to settle after the liner was installed, requiring a
geomembrane that could elongate in response to settlement. Also, a PVC liner could be
placed directly on the soft lake sediments, whereas extensive subgrade stabilization or
removal of the soft lake clay would be required to prevent bridging beneath other
geomembranes with smaller elongation properties at an unacceptably high cost to the
CSDPR.

Photograph 1. Lake bed soil conditions at Prospect Lake

The lining system selected for Prospect Lake consists from bottom to top of a 30 mil
PVC geomembrane placed directly over the prepared lake bed soils, an 8 oz/square yard
cushion non-woven geotextile, and cover soil mined from the lake bed, which varied in
thickness from 12 inches in areas inaccessible to the public to at least 2 feet thick in areas
that would be accessible by the public. In areas of soft clay or sediment, biaxial Tensar
BX1100 geogrid was used to improve subgrade support to allow for compaction of cover
soil placed above the PVC geomembrane to at least 92% of the maximum Standard
Proctor dry density.

Liner Construction
The construction team consists of: Watersaver Co., Inc. fabricating the PVC
geomembrane; Scientific Site Construction, Inc. (SSCI) installing all of the
geosynthetics; Mullett’s Excavating, LLC performing clearing, subgrade stabilization,

2
and cover soil placement above the liner; John Bowman Inc. serving as general
contractor; and K+A providing full-time quality assurance observation and material
testing for the CSDPR. Watersaver factory fabricated 113 PVC geomembrane panels
with a size of 70’ x 300’ and shipped the sealed panels to the project site (see Photograph
2). This resulted in Watersaver fabricating about 2.4 million square feet of PVC
geomembrane for this project. The factory fabrication was accomplished with over 64
miles of factory solvent-welded seams. The panels are sealed and shipped in plastic to
protect the liner from dirt, ponding water, and UV exposure during shipping and storage.

SSCI began liner installation in a freezing rain in late April and completed the project in
mid July with PVC sheet temperatures exceeding 170 0F. This resulted in a wide range
of geomembrane sheet temperatures for the air channel testing of field seams discussed
subsequently. Prior to site grading, the lake bed was combed to remove glass and other
debris that could mix with cover soil mined from the lake and placed above the
geomembrane. Prior to placing the liner, the subgrade was compacted and rolled with a
smooth drum roller, and inspected by both the liner installer and K+A for sharp objects.
Approximately 6.2 acres of geogrid was used to stabilize the softest lake areas. The
geogrid was placed directly on the subgrade and covered with 18 inches of on-site, clayey
to sandy on-site material. The shallow ground water was effectively dewatered to at least
4 feet below the lake bottom by excavating a longitudinal trench, and interconnecting
lateral trenches to a sump where water was removed by pumping, and the wide trenches
were subsequently backfilled to provide construction pathways to the softest areas of the
lake.

Photograph 2. Sealed PVC panels stockpiled at Prospect Lake

The large PVC panels were quickly unfolded and deployed (see Photograph 3). The use
of large panels reduced the number of field seams by 60% over a polyethylene
geomembrane alternative. During the construction phase, weather conditions, and the

3
earthmoving contractor’s ability to prepare the subgrade dictated the pace of
geomembrane installation and seam testing. SSCI personnel successfully placed, dual
track hot air welded, and air channel tested nearly all of the longitudinal and butt seams.

Photograph 3. PVC panels being unfolded and deployed

Air Channel Testing of Field Seams

A dual-track field seam was specified as the primary seaming method for the project.
Given the high cost of treated water used to fill the lake, a seaming process that allowed
for testing the entire length of the field seams was sought which resulted in the use dual-
track welds with air-channel testing. The air channel testing of PVC field seams is
gaining popularity and provides a number of advantages over destructive testing of PVC
seams (see Photograph 4). One of the advantages is the air-channel pressure can be used
to verify the PGI specified seam peel strength (PGI 2004) of 2.6 N/mm (15 psi) using the
sheet temperature and a relationship presented by Stark et al. (2004). Thus,
demonstrating that the air channel holds the required air pressure can be used instead of
destructive seam sampling and testing. Air-channel testing was challenging on this
project because the sheet temperatures varied from 400F to over 1700F in response to air
temperatures that varied from about 40 0F to 100 0F. In addition, the sheet temperature
occasionally varied over 100 0F in one day. SSCI also conducted a visual examination of
the inflated seams, and performed air lance testing along the exposed outside edge of the
dual seam to identify any seam defects or possible burn holes caused by wedge-weld
seaming.

The project specifications initially required destructive field seam tests every 1,000 feet
of field seam, but allowed the testing frequency to be reduced to one per 2,500 feet based

4
on successful air channel test results. This destructive sampling is significantly less
frequent than traditional destructive tests that are conducted every 500 lineal feet of field
PVC geomembrane seam. Initially air channel blowouts occurred during air channel
testing at temperatures exceeding 120 0F because of difficulties in controlling the air
channel pressure under high sheet temperatures that varied significantly during the test.
Specified burst pressures are currently provided only for sheet temperatures up to about
120 0F (Stark et al., 2004). To overcome this limitation, SSCI was allowed the option of
using air lance testing in place of air channel testing for sheet temperatures that exceeded
120 0F. This was done to limit damage caused by blowouts at higher temperatures where
the required burst pressure could only be extrapolated. Additional testing is being
conducted to extend the air channel test relationship in Stark et al. (2004) to sheet
temperatures greater than 120 0F.

After implementing the changes described above, the number of blowouts during air
channel testing was dramatically reduced. This allowed the frequency of destructive
testing to be reduced to one destructive sample per 2,500 feet of lineal field seam a 60%
reduction in the frequency of destructive testing required for the project, and an 80%
reduction over that normally required for field testing geomembrane liners. Photograph 5
shows two patches for destructive samples of T or butt seams to ensure an adequate field
seaming in the presence of a butt seam. A butt seam occurs when two fabricated panels
butt against each other, which usually involves seaming over a pre-existing weld. No
problem was encountered wedge welding the T or butt seams on this project.

Another advantage of air channel testing is the flexible PVC geomembrane allowed
people to visually observe the inflation of the air-channel as the air pressure migrated
along the seam and inspect the entire inflated seam for aneurysms or other seam defects.
Photograph 6 shows an inflated field seam and there are no defects in the welds because
the inflated air-channel exhibits a uniform shape, i.e., no aneurysms. In addition, the
entire length of field seam was tested using the air-channel test procedure to evaluate the
integrity of the dual seams, and air lance testing was used to identify possible defects
along the exposed outside edge of the dual seams.

Difficulties were also initially experienced in maintaining a constant seaming speed due
to uneven ground conditions beneath the seam, often resulting in seam burns. SSCI
solved this problem by folding back the edge of the liner being seamed by about 3 feet,
allowing the seam to rest on the underlying fold, which helped to bridge over uneven
ground.

5
 Photograph 4. Two PVC panels being hot air dual-track wedge welded in the field

Photograph 5. Only two destructive samples for T or butt seams which are covered by
the two patches shown

6
To date this is the largest use of air-channel testing for PVC geomembranes in the U.S.
Approximately eight miles of field seams were created and air-channel tested during this
project. Over 64 miles of factory seams were used and only 8 miles of field seams.
Thus, about 89% of the seams were created in a factory-controlled environment while
only 11% were created in the field.

Photograph 6. Air-channel test of a PVC field seam in cool weather and the inflated
seam is still readily visible

Multi-Tasking with Air Channel Testing

In the liner system, the geomembrane is overlain by about 2.6 million square feet of non-
woven geotextile. Installation of the geotextile was accomplished using 573 rolls of
geotextile with dimensions of 15 feet x 300 feet. The geotextile edges were tacked
together using a heat welder, totaling over 34 miles of tack seams, a labor- and time-
saving alternative compared to temporarily anchoring the geotextile using sand bags prior
to placing the cover soil. The rate of geotextile deployment and seaming was greatly
facilitated by the air channel testing of the PVC geomembrane seams. The air channel
testing allowed the geotextile to be deployed immediately after geomembrane
deployment instead of waiting for test results on destructive samples, with the exception
of areas where limited destructive testing was performed. Photograph 7 shows the
geotextile being deployed over a PVC panel while the panel is still being welded.

7
Photograph 7. PVC panels being seamed will the overlying non-woven geotextile is
being deployed

Summary
Prospect Lake in downtown Colorado Springs was refilled and the city is once again
enjoying the beauty of this wonderful urban treasure as evident by the Labor Day 2005
festivities including a hot air balloon festival (see Photograph 8). The use of a PVC
geomembrane facilitated the saving of Prospect Lake because the problematic lake bed
soils did not have to be excavated, treated, or improved significantly to accommodate the
liner. In addition, the liner installation was completed much quicker than if a rigid
geomembrane requiring more field seams and greater subgrade preparation was used,
which facilitated completing the project in a little over three months, which allowed the
lake to be filled in time for the Labor Day festivities. The use of air channel testing
resulted in a significant reduction in destructive testing and greater assurance of seam and
liner quality. However, both air channel and air lance testing were used at sheet
temperatures greater than 1200F to ensure identification of defects along the exposed
outside edge of the dual seams.

Prospect Lake represents the largest U.S. PVC geomembrane installation to date that has
utilized air-channel testing and limited destructive testing. It is anticipated that
successful projects like Prospect Lake will increase the use of non-destructive tests to
enhance the quality of completed liners and the use of PVC geomembranes.

8
Photograph 8. Hot air balloons over refilled Prospect Lake for the 2005 Labor Day
weekend

REFERENCES
PVC Geomembrane Institute (PGI), 2004, “PVC Geomembrane Material Specification
1104”, University of Illinois, Urbana, IL, www.pvcgeomembrane.com, January 1, 2004.

Stark, T.D., Choi, H., and Thomas, R.W., 2004, "Low Temperature Air-Channel Testing
of Thermally Bonded PVC Seams," Geosynthetics International Journal, Industrial
Fabrics Association International (IFAI), Vol. 11, No. 6, December, pp. 296-310.